Does Apple's new foundation models include a Vision API for accessing on-device LLM capabilities?

Question

Created 1w

Replies 2

Boosts 0

Participants 3

I couldn't find information about this in the documentation. Could someone clarify if this API is available and how to access it?

Answered by DTS Engineer in 842599022

If you are working with computer vision, you might consider using other APIs. For example, for image generation, Image Playground should help.

The Apple Foundation Models introduced in WWDC25 accepts text as an input and generates text as an output. If your input is not text, depending on your concrete use case, you might consider converting your input to a text description, which the models should be able to handle.

Best,
——
Ziqiao Chen
 Worldwide Developer Relations.

Boost

Answer 1

Apple Designer OP

Apple

1w

To clarify, are you looking for computer vision, image creation, or visual understanding on-device? Or are you looking for on-device LLM access for visionOS?

1

Answer 2

DTS Engineer OP

Apple

1w

Accepted Answer

If you are working with computer vision, you might consider using other APIs. For example, for image generation, Image Playground should help.

The Apple Foundation Models introduced in WWDC25 accepts text as an input and generates text as an output. If your input is not text, depending on your concrete use case, you might consider converting your input to a text description, which the models should be able to handle.

Best,
——
Ziqiao Chen
 Worldwide Developer Relations.

2